AMP: The web’s first steps into the Performance Age
We are witnessing an evolution of the mobile web. Google’s Accelerated Mobile Pages (AMP) have driven a renewed focus on web performance, bringing with them a change in user expectation that will topple the foundations of Responsive Web Design. What factors led to the rise of AMP and how will web design change as we enter this new age of performance?
In 2016, Google begin offering links to validated, cached versions of web pages in its search results, identified by a lightning bolt symbol. These pages load much faster than traditional mobile web page, 85% faster according to Google.
Results are so far suggesting big payoffs for early adopters of AMP pages, with some companies benefitting from 70–100% lifts in conversion and traffic following the launch of AMP.
Since then, statistics have been flying in that indicate just how big an effect load times can have on a website’s success. Google states that 53% of users who click on a link in their search results will abandon the site if it takes more than 3 seconds to load. If that search result leads to an online store generating $100,000 per day, a 1 second delay loading the first page will cost the merchant $2.5 million in lost sales every year.
The web performance timeline
It’s easy to see why companies want AMP; it gives their mobile web users almost instant access to content, making it more likely they’ll continue using the site and ultimately “convert”. But what has led to this intolerance of sub-par performance?
Mid-2000s: Early Smartphones
The Blackberry and other early smartphones began to offer mobile internet to the masses in the early-mid 2000s. Limited computing power and slow connections meant that laggy and cumbersome experiences were expected. Needless to say, users were slow to catch on.
2008: Apps
The mobile web really hit the mainstream with the introduction of the iPhone and the debut of The App Store. Installable apps introduced a narrowed focus on task completion, their remit was to perform simple tasks faster and with less intrusions traditionally experienced on the mobile web. Users got used to shorter load times and mobile-optimized designs. These would later become the guiding principles of AMP.
2010: Tablets & Responsive
Smartphones, tablets and “phablets” were flooding the market. Companies needed a way to deliver acceptable experiences on all these screens without blowing their budgets developing native apps for multiple platforms. Responsive Web Design (RWD) was born out of this growing desire for a cost-effective alternative to apps and a need to meet increasing user expectations for tailored experiences — no matter which device they chose to use.
2014: ‘Mobile-friendly’
While Responsive Web Design was helping the web catch up to this rising user demand for mobile-optimized layouts, Google began introducing the “mobile-friendly” label to their search results. It let users know the site they were about to visit had been optimized for small screens, either through RWD or an alternative framework.
2016: Progressive Web Apps & AMP
Widespread adoption of RWD has caused Google to retire the Mobile-friendly badge, at the time citing that 85% of its search results were now mobile-optimized. Users expect layout optimization at a minimum, and the sites that surpass their expectations are those that are fastest to load and easiest to use. Accelerated Mobile Pages and Progressive Web Apps (PWAs) have emerged to meet this evolved user demand for app-like performance on the web.
Responsive is dead. Long live performance!
Responsive Web Design played a key role in getting mobile optimization to where we are today, but it is not set up to deliver in this new age of high performance. The approach of “one codebase for all screens” made mobile development accessible for all businesses, but most didn’t consider the consequences of careless implementation. Desktop-sized payloads were being squeezed to fit smaller processors and bandwidths without proper performance considerations, creating an influx of slow and overly complex mobile experiences.
Another contributor to poor performance is the abundance of tools on offer today that help websites deliver content to their users. These tools often use heavy scripting languages such as JavaScript to pipe images, fonts, ads and social content from external sources into the website. These requests stack up and impact performance, especially on responsive sites where mobile systems are forced to accommodate cumbersome scripts, often useless to the mobile user.
Small screens need a tailored experience, not only in the design of the page, but also in the code under the hood. This renewed focus on reducing code complexity to improve load times epitomizes AMP. Companies are now looking outside of the RWD status quo to solutions like AMP and PWA to offer their customers a mobile experience that meets their performance expectations from start to finish.
Is AMP a good thing?
The AMP Project has attracted its fair share of critics because of the way it works. It requires websites to render their content in a subset of Google-approved HTML, using a set of Google-developed tools, to appear on a Google URL. A few high profile developers have published scathing criticism of AMP, calling it “anti-web” and suggesting a more sinister desire coming from Google to control other people’s content.
Whether you share this skepticism or not, you can’t deny that AMP is giving users what they want. It meets their demand for instant access to content, and the average user could care less that they’re accessing it on a Google domain.
Where is AMP leading us?
As more and more AMP pages make their way into our search results, it’s bound to spark a renewed focus on site speed as companies try to get ahead in the race. We may see a shift in the one-codebase-for-all approach, with web developers taking more control of where and how content is served under constrained connections.
This pivot from the layout optimization practices of Responsive Design to performance optimization under AMP, can only be seen as a welcome step forward in the eyes of any mobile web user. The standards are being reset and the rest of the web will look to follow AMP’s example.
An announcement from The AMP Project in March 2018 appears to recognise this imminent emergence of non-AMP options for delivering instant content:
“We are taking what we learned from AMP, and are working on web standards that will allow instant loading for non-AMP web content… It will be just one of many choices, but it will be the one we recommend.” Source
It seems that rather than claim a monopoly on web frameworks, Google aims to position AMP as the leading opinion on what makes a highly performant web experience. This should create open competition and may well quash the sceptic’s claim that the high performing sites of tomorrow are only accessible from a Google search.
A look to the future
Up to now, the focus has been on improving mobile performance, which makes sense given that mobile connections and processing power are more susceptible to slow performance than desktop devices. However, there is no reason why AMP could not be applied to larger screens. This direction is reflected in PWA development, with a shift in focus to bring tailored app-like experiences to the desktop.
Web development will catch up and begin providing the kind of load times we’re seeing with AMP. High performing sites forming the majority of search results will render the AMP badge obsolete. It will be a case of history repeating itself — like when Google made its Mobile-friendly badge redundant in the wake of Responsive Web Design.
Perhaps then the race will be to provide not only the fastest experience but also the most enjoyable experience, bringing user experience design back to the forefront.
Conclusion
Personally, I wish AMP didn’t have to exist. I wish mobile performance never reached the point where Google-hosted HTML was the only way to provide users with acceptable load times. I hope The AMP Project is the catalyst that galvanizes website owners, web developers, and software companies to step back and really consider how their decisions affect performance–and ultimately, user experience. The AMP Project’s recent announcement appears to recognize this imminent catch up, where the instant load times we’ve come to expect are delivered with or without first performing a Google search.
As a web designer and user, I’m excited about entering the Performance Age. It will no doubt impact the web in a big way, and high performance will not stop at mobile or at simply measuring page load. Quicker load times are just one factor in delivering good user experience. The prospect of instant loading plus intuitive, enjoyable experiences is something I can’t wait to design and use myself.
